Karanis: the Virtual and the Reality

The 7th International Symposium on Virtual Reality, Archaeology and Cultural Heritage VAST (2006) M. Ioannides, D. Arnold, F. Niccolucci, K. Mania (Editors) Short Presentations VR Modeling in Research, Instruction, Presentation and Cultural Heritage Management: the Case of Karanis (Egypt) W.Z. Wendrich,1 J.E.M.F. Bos,2 K.M. Pansire1, 3 Department of Near Eastern Languages and Cultures, University of California, Los Angeles, USA 2 Past 2 Present-ArcheoLogic, The Netherlands 3 The authors would like acknowledge contributions by Terry Wilfong (Kelsey Museum of Archaeology, University of Michigan), Lisa Snyder (Experiential Technology Center, UCLA), Zoe Borovsky (UCLA Digital Humanities Incubator Group), and Zahi Hawass, secretary general of the Egyptian Supreme Council of Antiquities. 1 Abstract Since Fall 2005, a team at the Experiential Technology Center of the University of California, Los Angeles (UCLA), and archaeologists from the UCLA and the Rijksuniversiteit Groningen (RUG) expedition to Fayum have been creating a Virtual Reality model of the ancient town of Karanis. The model has multiple purposes, perhaps the most innovative being its future use as a management tool for the archaeological site. The model will aide in monitoring the decay of the town since the early 20th century, and in reconstructing the buildings, but it will also be used to study aspects of routing and the use of space, as well as a way of explaining architecture, principles of stratigraphy, and life in the ancient town to students. These seemingly disparate aspects of the VR model all aid the compilation of a site management plan for Karanis. This short paper presents a work in progress, envisioning the full potential that 3D technology implies. Categories and Subject Descriptors (according to ACM CCS): J.3 [Arts and Humanities]: Architecture From 1924 to 1935 the University of Michigan excavated a large part of Karanis, concentrating on the town’s central quarter. The excavations revealed a large mud brick town, with two stone temples, multiple storied houses, granaries, dovecots and ample evidence of grain and olive processing [Boa33; BBP31; Hus79]. The excavations were a race against time, with the purpose to save and document as much of the town remains as possible in the little time they had, before the sebakhin would excavate all. Before the Michigan excavations finished, the destructive work of the sebakhin was brought to a halt, by urgent claims that the site represented an important historical monument. The excavations cleared a wide area, to enable an in-phase overview of the different strata. Late Roman layers (dated to the third to fourth centuries CE) were removed to understand the earlier periods of occupation. After excavation was finished, in 1935, the mud brick monuments were mostly left open to the elements. 1. Introduction In the late 19th century three British scholars, Grenfell, Hunt and Hogarth visited the Fayum oasis in search of scholarly treasure (Figure 1). They were after Greek papyri, which were found in great quantities by the socalled sebakhin, farmers who were excavating ancient settlements to use the mud brick on their fields as fertilizer. The papyri and other antiquities were a lucrative by-product of these activities, and became a focus of the antiquities trade. The three British gentlemen had no interest in architecture, or archaeology, although they did remark on the exceptional preservation of wooden objects at the Fayum sites, in particular at Karanis. After two thousand years agricultural tools, doors, windows, roofs, textiles and baskets were still found intact and in amazingly good condition, but their focus was the papyri [GHH00]. The architecture was left to be cut away by the sebakhin. In 1924 the University of Michigan started excavations in Karanis, after a complicated battle with the sebakhin, who by that time were well organized and had a little rail line running from the center of the ancient town towards the edge, so that the mud brick was harvested on an industrial scale. The maps and elevations of the Michigan expedition show at least six major occupational phases, but also a large area, encompassing the entire center of the town, marked ominously “Area Entirely Destroyed By Sebakhin”. 2. Present situation of Karanis The town of Karanis was probably founded in the third century BCE, during the efforts under Ptolemy II Philadelphus to bring the Fayum under cultivation. These earliest (Ptolemaic) levels were not reached by the Michigan excavations. A recent re-interpretation of the archaeological remains puts the end of Karanis to the sixth or seventh century CE [Pol98]. These nine 2 W.Z. Wendrich, J.E.M.F. Bos & K.M. Pansire / VR modeling, the case of Karanis (Egypt) and west, and continue at the west side of the main road from Cairo to the Fayum (Figure 2). The site east of the road is open to the public and offers tourists an open air exhibit (the excavated part of the site, Figure 3), and a small museum which boasts antiquities from all periods of Egyptian history, from pottery dated to the Predynastic (3000 BCE) to the porcelain of Egypt’s last King Farouk (1950’s CE). The museum does not focus specifically on the Fayum, although several objects on display have been excavated in the Eastern part of the Fayum (Hawara). The museum was established in the 1960’s as part of a philosophy to give all Egyptians the opportunity to find a broad overview of their national history in locally accessible museums. At present, the open air museum at Karanis is a bewildering experience for the visitor. Sand-filled pathways lead through the excavated areas, and in many places run over underlying structures, exposing older, vulnerable mud brick walls. To the uninformed eye, an enormous deep vacant space in the center of town, flanked by two stone temples, seems to represent a curious type of inversed agora. Without knowing the history of exploitation of the site, the town’s fabric is incomprehensible. The mud brick buildings, excavated by the Michigan expedition, show evidence of rapid decay: Figure 1: Location of the Fayum, the study area of the UCLA/RUG excavation project and the ancient town of Karanis. centuries of occupation have created a depth of deposit which in some places is over ten meters. The large area excavated by the Michigan team, and the part destroyed by the sebakhin still cover no more that 25% of the entire site. On satellite photographs it is clear that the unexcavated parts of the town extend to the north, south Figure 2: Google Earth photograph of the ancient site of Karanis, with the approximate boundaries of the ancient remains, the cemetery west of the Cairo-Fayum desert road, and the destroyed central part of the site between the North and South Temples. W.Z. Wendrich, J.E.M.F. Bos & K.M. Pansire / VR modeling, the case of Karanis (Egypt) 3 Figure 3: Center of Karanis looking south west towards the South Temple. The low lying area between the mud brick building and the stone temple is the “area entirely destroyed by sebakhin” (see also Figures 2 and 4). severe undercutting by wind erosion, dissolving of the plaster and bricks by rare but occasional precipitation. Conservation efforts of the Supreme Council of Antiquities and a French team in the 1970’s have concentrated on a few buildings: the North and South Temple and a bath house in the west part of town. shipped to Michigan and are now in the Kelsey Museum in Ann Arbor. Another portion of the Karanis finds (botanical remains, but also agricultural implements) are on display in the Agricultural Museum in Cairo. The botanical research, headed by the co-director of the Fayum project Dr. René Cappers, studies these finds, but also tests rigorous sampling methods in new excavations, to determine how representative the present Karanis botanical collections are. Archaeology has seen a development in recent years, from purely research based projects, to an approach that takes much more responsibility for site preservation and presentation, as well as information to the general public. It is in this context that the UCLA/RUG Fayum project, directed by Willeke Wendrich and René Cappers, started to collaborate with Jolanda Bos, from the Dutch archaeological company Past2Present-ArcheoLogic, to investigate the possibilities of site management. The project has received support from the Antiquities Endowment Fund, administered by the American Research Center in Egypt for the first phase of the work: the initial evaluating phase of the management to be summarized in a position paper. 3. Environmental archaeology in the Fayum area In 2003 an excavation team of the University of California, Los Angeles (UCLA) and the Rijksuniversiteit Groningen (RUG), started a landscape project in the northwestern region of the Fayum. The project focuses on land and water use in the development of agriculture, from prehistory to present. The project’s concession area includes extensive Prehistoric remains and three Greco-Roman settlements: Qaret Rusas, el Qarah el-Hamra (discovered in 2002), and Kom Aushim / Karanis. The Prehistoric remains provide information on the earliest agriculture in Egypt, the cultivation and storage of wheat and barley. A study of Karanis is undertaken to augment information provided by the Michigan excavations. Many of the botanical remains have been 4 W.Z. Wendrich, J.E.M.F. Bos & K.M. Pansire / VR modeling, the case of Karanis (Egypt) 4. Karanis site management plan The remains of the ancient town of Karanis were not adequately protected, through backfilling after excavation was finalized. Since the excavation of the University of Michigan, the mud brick town has been exposed to the elements. A rough estimate of the decay in the last century leaves the impression that more than 50% of the exposed remains have eroded in the past seventy years. For the UCLA/RUG expedition it was apparent that the monuments needed to be dealt with in some way. Excavation is not the main focus of the team, which feels a responsibility to aid the preservation of the remains of Karanis for future generations. As became clear from an initial survey, complete conservation of the site would be an impossible undertaking, both technically and financially. Difficult choices need to be made regarding the conservation of this vast area. To provide well founded grounds for these decisions, the site needed to have a management document. A team, headed by Jolanda Bos, is developing a site management plan to safeguard the preservation of the remains, at the same time enabling well substantiated decisions regarding access, information, conservation and future research. For this purpose the determination of the value of the Karanis monuments and the site as a whole will be assisted by developing several management assessment tools. These will also provide a toolkit for the archaeologists working at the site, in relation to site and heritage management in Fayum. Instruments, needed to determine the direction of the archaeological heritage management, will be developed in a bottom-up approach. At the core will be the archaeological remains and the interests of the different stakeholders, in contrast with an approach in which ‘best practice’ is implemented without a thorough analysis of a site’s potential. One of the main assessment tools for (future) archaeologists and heritage managers of Karanis is a survey of the state of conservation and rate of decay of the mud brick buildings. Important elements in the study are the publications, maps, excavation notes and other documentation of the previous archaeological expeditions. During our initial year at the site, it became apparent that one of the sources to assess these issues are the photographs taken over seventy years ago by the Michigan expedition. Comparing these with documentation of the modern state of preservation should facilitate answering questions such as: How fast does the mud brick deteriorate? What has disappeared during the past seventy years after exposure of the buildings? What are the processes that are causing this? To aid us with the assessment, and to enlarge the impact of our findings, we decided to make use of the results of a parallel project, the creation of a virtual reality model of Karanis for use in the class room at UCLA. 5. Virtual Reality model of Karanis In Fall 2005 a start was made to create a real time Virtual Reality model of the town of Karanis (Figure 4). With funding from the Office of Instructional Development at UCLA, three graduate students of the Department of Near Eastern Languages and Cultures at UCLA, Kandace Pansire, Eric Wells, and Carry Zarnoch are rendering the plans and elevations of the Michigan publications into a three-dimensional model of the town. The work is done in the laboratory of UCLA’s Experiential Technology Center, and greatly assisted by the center’s Associate Director, Lisa Snyder. The rendering program used is Multi-Gen, which enables the creation of a real time model, which is generated on the fly. The model is created in such a way, that it serves multiple purposes, five of which are highlighted below. Figure 4: First stage of the Karanis VR model development), based on published maps from University of Michigan publication. In the forefront map indicates the “area entirely destroyed sebakhin”. (in the the by 5.1. The Karanis VR model as an instructional tool The values of VR modeling in archaeology is generally recognized [Bar00]. Students of architecture or archaeology learn how to read maps, plans and elevation drawings. Students of history, art history, or ancient languages often do not have this background and show great reluctance to study information in a nonlinear or non-narrative way. Understanding an ancient city, the activities of its inhabitants, the role of neighborhoods, location of industrial quarters, routing, social gathering spaces and all other aspects that are part of a town’s fabric, requires a spatial approach to the subject. By creating the VR model, the plans and elevations are translated into a three dimensional representation, familiar to a generation accustomed to computer games. Perhaps the most important aspect of the instructional model is the display of phasing of the site. Rather than painstakingly studying a series of eight large fold-out maps, the student can toggle between phases. This enables an understanding of urban development, re-use of buildings, re-definition of city quarters, and concepts of growth, decline and abandonment. W.Z. Wendrich, J.E.M.F. Bos & K.M. Pansire / VR modeling, the case of Karanis (Egypt) 5 5.2. The Karanis VR model as a research tool The potential of the Karanis VR model as a research resource is as enormous as that of the teaching tool. Complex phasing of individual buildings or even segments of buildings, can be integrated in the phasing of the site as a whole. The model can be used to summarize and integrate micro-analysis into an overall stratigraphic overview. Secondly, the virtual reconstruction of buildings or entire sections of the town will enable testing of hypotheses ranging from building methods, to inside light levels, temperature control, storage capacity and lines of vision. Architectural reconstructions have an important heuristic function, because decisions on, for instance, wall height, roof lines, and stair wells require decisions on very specific structural details. Each choice has consequences which are difficult to foresee when made on paper, while the virtual reconstruction translates each alternative into a very specific set of delimiting conditions. A third example is a research project starting in Fall 2006 in which the model will be used to represent the spatial distribution of grinding stones and olive presses. The site of Karanis is littered with large round limestone milling stones, smaller granite hand mills, heavy half-rounded press weights and large stone press foundations. Although not all of these are in situ, a study of the distribution is expected to result in the identification of centers of activity of grain processing and olive oil production. In combination with a detailed study of each grinding or mill stone and the wear marks on these, reconstructions of the different production types will be made. By connecting the VR model to a GIS, full use can be made of the three dimensional spatial analytical functionality. extant remains can be compared to the recreated model. Such on-site information panels also aid the visitors in understanding how the bewildering mud brick landscape ahead represents different phases in town use, partly dug away by excavators and looters. The use of VR models has been applied with success for this purpose at for instance the sites of Ename and Luang Prabang [PCK*00; Let99]. One of the aspects underdeveloped in archaeological information allocation to the general public is the methods archaeologists use. Popular on-site presentation of archaeology often still focuses on artifacts, rather than the archaeological context. The public is, therefore, often not aware of the potential information to be gained from an archaeological site, and the modern excavation and scientific methods used to do so. In addition to explaining the history and development of the town, the VR model can be programmed to show the visitor a glimpse behind the scene: it will present the research potential of Karanis as an archaeological site, the vulnerability of the remains, the efforts to preserve the site, and the importance of public awareness and participation. The model will, therefore have a prominent role in the Karanis site information center. 5.5. The Karanis VR model as an assessment and management tool At several stages the VR model for the site will serve in the management of the ancient town. Its function as threat monitor and explanatory aid were outlined above. In addition to the educational aspects, the model will be used to design the routing for visitors. At present tourists are allowed to roam freely on site, while there is no discouragement to climb on buildings. Assessment of areas under immediate threat can be monitored closely, and the results conveyed to the public version of the model, to be used as a vivid illustration why the vulnerable mud brick should not be touched. Based on the detrimental stress, selected visiting highlights and routing, a selection will be made of areas which require immediate backfill, or other consolidating interventions. Not all remains of the town of Karanis were excavated by former expeditions. In fact a considerable part of the site still lies hidden in its protective matrix of sand. These unexcavated areas which are not under direct threat of deterioration by natural or human factors, should be preserved in situ, rather than endure ex situ conservation. The results of surveys with nondestructive techniques, such as a magnetic survey to be done in Fall 2006, will be added to the VR model to extend the town to its full former size. At that stage, the model can be used to implement a site zoning as a management tool for the different potential usages. Areas that are completely off-limits, public areas restricted to tourist routing, increased access for guided specialized trips, and research areas are some of the potential zones to be defined. 5.3. The Karanis VR model as threat monitor During the Fall 2005 field season, Bastiaan Seldenthuis, the photographer of the project systematically recorded the wall faces marked on the old University of Michigan Karanis site plans. This time consuming project, which will be finalized in the Fall 2006 field season, enables us to correctly project the wall faces on the Karanis VR model. By underlying the 2005/2006 photographs with those made in 1924-1935, the VR model will enable us to visualize the wall decay during the past seventy years. Similar photographic recording of all, or part of the standing walls on site in future years will provide one method of documenting the state of preservation and an instrument to monitor further deterioration. 5.4. The Karanis VR model as presentation tool The VR model will be part of an exhibit that will elucidate the function, appearance and history of the town. Similar functionality used to teach the ancient urban fabric to students, will be used to guide tourists through the virtual town as preparation of a visit to the actual remains. Stills from the model will present reconstructions of particular quarters on site, so that the 6 6. Conclusion W.Z. Wendrich, J.E.M.F. Bos & K.M. Pansire / VR modeling, the case of Karanis (Egypt) Originally designed as an instructional and research tool, we are just now exploring the full potential of VR modeling technology by making it accessible to all the stakeholders of a multi-faceted project such as the Karanis site. The model’s instructional value can be utilized as a wake-up call for policy makers when threats and decay are made shockingly visible by overlaying photographs from different periods; visitors, when confronted with this evidence, may comply more willingly with limited access if they understand that the rules are designed to preserve what previously was invisible to them. Research results feed into public information as well as management decisions. Assessment of the current situation, aided by a threedimensional spatial representation can go hand in hand with a representation of the foreseen improved state of affairs after specific interventions. The VR model can be used as a record, a representation, a prediction, and a test of previous, current and future situations. The potential of VR in cultural heritage resource management is vast, but it is our task to envision the new capabilities it gives us, i.e., its potential to enable research to proceed, while preserving the site itself. Most importantly we must avoid designing our projects and predicting our results based on existing technologies rather than taking into account the full potential that this new 3D technology implies. The Karanis VR model, with its easy accessibility, may become the center of focus for students, archaeologists, visitors, local population, government oversight officials and policy makers. It will bring together the important stakeholders of the Fayum heritage and cover the multiple angles of the Karanis heritage management. [Let99] Letellier, R., Virtual Reality? A new tool for sustainable tourism and cultural heritage sites management. Proceedings CIPA Symposium (1999). [PCK*00] Pletinckx, D., Callebaut, D., Killebrew, A.E., Silberman, N.A. Virtual-reality heritage presentation at Ename, Multimedia, IEEE (2000), 45-48. [Pol98] Pollard, N. (1998) The Chronology and Economic Condition of Late Roman Karanis. An Archaeological reassessment. Journal of the American Research Center in Egypt 35 (1998), 147162. References [Bar00] Barceló, J.A., Visualizing what might be, an introduction to virtual reality in archaeology, Virtual reality in Archaeology, Barceló, J.A., Forte, M., Sanders D. (Eds.), 2000, 9-36. [Boa33] BOAK, A. E. R. (ED.) Karanis, the Temples, Coin Hoards, Botanical and Zoological Reports Seasons 1924-31. Ann Arbor: University of Michigan Press, (1933). [BP31] Boak, A. E. R. and Peterson, E. E. (Eds.) Karanis, Topographical and Architectural Report of the Excavations During the Seasons 1924-28. Ann Arbor: University of Michigan Press, 1931. [GHH00] Grenfell, B. P., Hunt, A. S. and Hogarth, D. G. Fayûm towns and their papyri. London: Egypt Exploration Fund, 1900. [Hus79] Husselman, E. M. Karanis, Topography and Architecture. Ann Arbor: The University of Michigan Press, 1979.